Is Thomas Livermore Trustworthy?: A Story about Memory, Memoirs, and the Civil War

Hugh Dubrulle Blog Post

I recently completed Thomas Livermore’s Days and Events 1860-1866 because it is, to my knowledge, the only memoir written by a veteran of the 5th New Hampshire Volunteer Infantry. Livermore served with the regiment from its inception, starting as a 17-year-old 1st Sergeant and working his way up to Captain before obtaining a series of staff appointments, mainly with the II and XVIII Corps. If Days and Events 1860-1866 is not a classic of Civil War memoirs, it is still an informative and enjoyable read that reveals much about its author and his regiment. For that reason, it has become a key primary source employed by those who have done research on the 5th New Hampshire.


As I read the book and came to understand how Livermore wrote it, however, I asked myself, “Is it reliable?” Livermore produced the manuscript between 1867 and 1872, scribbling away when he had free time on Sundays. Since he had not kept a diary during the war, he worked from memory. In the years after he completed the draft, he revised it based on books he read as well as conversations he had and correspondence he exchanged with other veterans. That reliance on others colored his impressions, especially when he sought to recollect his years as a staff officer.


The passage that provides perhaps the best example of the dangers associated with this mode of proceeding appears in his account of Gettysburg, where he served as the staff officer in charge of the II Corps’ ambulances. Livermore could not remember exactly where he was, what he did, and what he saw during Pickett’s Charge, largely because his time was divided between overseeing the ambulance train and serving as an aide to Major General Winfield Scott Hancock, his corps commander. He wrote that his duties “carried me to every point where our corps was engaged” and “I was obliged to ride constantly to and fro.” For that reason, he wrote, “I have not the benefit of the sequence of events which would have been witnessed by me if serving as an aide all the time.” This confusion, which left Livermore with a memory salad, made him susceptible to thinking that he saw what he may have actually learned after the fact. In recounting the events of July 3, 1863, he produced this revealing passage: “It seems to me certain that I saw the rebels come through our lines [near the famous copse of trees], but I can’t arrange the events chronologically, and I almost fear I have heard about the rebels coming through so many times, that I have blended these accounts with my intimate knowledge of the looks of our lines and made myself believe I saw them come through when in fact I did n’t.” This statement is all the more interesting because only two pages later, he states emphatically, “If I had been serving with my regiment [from which he had been plucked for staff duty only a couple of days before], I could have told just what I did at all times.”


What should one make of these statements? To what extend did Livermore understand the workings of his memory? As I struggled with these questions I chanced upon an article in the May 2021 issue of The Atlantic entitled, “You Won't Remember the Pandemic the Way You Think You Will.” This essay, which is based on current findings in psychology, makes four major points about memory. First, we tend to remember with some specificity the high and low points in our lives, with trauma in particular “engraving painful and long-lasting memories.” Second, these moments become “episodes” in our memory that are “invested with meaning.” They help us compose a “narrative identity” and “continuous sense of self” that helps define who we are. Third, as we experience an event, we compose “our memories in anticipation of sharing them,” a response that colors both our perceptions and recollections. In so doing, we weave events into a story. Fourth, in sharing this story with others, we begin to modify our recollections of what happened by borrowing parts of other people’s memories of the same event. We thus become members of a community engaged in a “crowdsourcing project of organizing collected memories,” and our recollections thus become collectively constructed.


How does this article help us understand Livermore’s memories? His account provides ample evidence that his mind worked in accordance with the findings in The Atlantic article. It is clear, for example, that Livermore composed his memories in anticipation of sharing them. Moreover, a clear narrative identity and sense of self emerges in his narrative. Finally, evidence from Days and Events 1860-1866 suggests that Livermore may have been onto something in thinking that his memories as a company officer were less susceptible to ex post facto collective construction than his recollections of time on various staffs.


On the one hand, while fighting with the 5th New Hampshire, Livermore was exposed to a great deal of combat trauma that made an indelible impression not easily misremembered. Indeed, his depictions of the major battles in which he fought with the regiment are not only lengthy and detailed but also dwell on gruesome incidents. Moreover, Livermore’s ambit was restricted to a small number of men who occupied a limited amount of space. His account of life in the regiment centered around his company which, after the Peninsula campaign, probably never numbered more than 30 men (and often much less). It was easy to construct a chronological account of such a small group, and almost all of its members were neither as articulate nor as fluent with the pen as Livermore was. In other words, this body of men—which constantly turned over due to death, discharge, and stays in the hospital—did not possess the wherewithal to challenge or influence Livermore’s view of events. 


On the other, as a staff officer, Livermore’s duties were variegated, and he was involved with many, many more men strewn over a much larger area. From this position, he obtained occasional glimpses of different faces of each battle or operation, leading to a disconnected mélange of impressions that he shaped after the war through reading and conversation. Moreover, he was surrounded by a number of educated officers on the staff who could contest as well as sway his view of his experiences. Finally, as a staff officer, he was exposed directly to far less trauma, and his duties were far less memorable.


Should the reader, then, treat Days and Events 1860-1866 with caution? Yes, because all sources require caution, especially those written well after the events they describe. But The Atlantic article on memory, when coupled with Livermore’s own insights about the workings of his memory, indicates that some sections of the book (e.g. Livermore’s recollections of service in Company K and then Company E in the 5th New Hampshire) are perhaps more trustworthy than others (e.g. his memories of staff work). 

8 Replies

Post Reply

Hi Hugh,
The question of how reliable an individual's memory is, particularly in relation to our use of sources as historians, is of especial interest to me, given both my project and my training. The "myth of Mary Bowser, Black Civil War spy in Jefferson Davis's white house" asserts she had a "photographic memory" (shouldn't that be daguerreotypic for period accuracy????) for anything she saw or heard. First of all, memory for what one hears functions differently than for what one sees/reads, so quality of one kind of recall is unrelated to the other kind of recall. Moreover, I have read enough documents by or about her to know that she certainly misremembered things. She also sometimes altered details in public speeches, private conversations, or personal correspondence, sometimes as part of a calculated rhetorical strategy and sometimes to protect herself from possible retaliation by white supremacists. Moreover, the original claim about her "photographic memory" is from a very dubious source who first wrote it down in the mid-twentieth-century, having claimed he heard it years earlier from an aunt who heard it from her father who told her on his deathbed about his wartime espionage and included this detail about "Mary Bowser." No corroborating evidence has been found that the speaker of the deathbed confidence was ever part of the network, and of course the third-hand rendering of his account would be questionable even if he were involved. But it's more likely the author of the mid-twentieth-century account made it up entirely.

As I've reflected on this "perfect memory" claim, I've come to believe that it also diminishes the actual accomplishments of the real Mary "Bowser" Richards Denman. She referred to herself as a detective -- a role that implies actively seeking information, piecing together evidence from different sources, and assembling a "case." To date, the specific examples I have of her doing so are from the postbellum period, when she was documenting white supremacist organizing and imploring government officials to protect the newly emancipated. But she certainly described herself as having been "one of Uncle Sams detectives" during the war, implying a far more active and important role than mere mechanical recall. Understanding that she undertook this kind of active role as an African American woman is especially important as we undo the (how is it still so prevalent???) myth that blacks were passive recipients of an emancipation given by Lincoln. Or that the Civil War was "men's business" with women on the sidelines.

Moreover, these questions of memory, of partial sources, of what we can and cannot piece together -- scholars of African American history have produced a great deal of insightful work in this area, precisely because of what might seem like a dearth of resources that center African American experiences, and that are not limited by being directed at white audiences or mediated by whites in how they were recorded or preserved. To be sure, understanding where to find these sources is important, but so is understanding how to interpret them. I will admit to feeling very validated last week when Jim Downs, during a SCWH panel, noted that his initial scholarly training was in African American literary criticism and theory. He cited many of the same influences -- Barbara Christian, Toni Morrison, Farah Jasmine Griffin -- whose scholarship I was immersed in as an undergrad and graduate student, back in the last century. In recent years, historians focused on African American/Black history have continued to extend the effort these women began to "read the silences, read the gaps, be attentive to the ellipses" (as Griffin put it).

This is not to say this approach isn't as important when the sources are white. Quite the opposite: the presumption that sources are accurate and unproblematic is not one any of us ought to make. But it's been a presumption made quite often across the history of American history. So while modern psychological studies/recent Atlantic articles may be one way to think through what we can or cannot learn from Livermore's account, you might explore how you can draw on scholarship that thinks about this in terms of race, particularly foregrounding the perspective of the enslaved (one place to start is this special issue of History of the Present )


Dear Hugh, 

Thank you for this thoughtful post.... a couple of quick thoughts that I hope are of some help... 

Does the Atlantic article suggest to us that the field of psychology has reached a broad consensus about what human memory is and how it operates, and that historians can draw on this consensus profitably (without each of us having to go through graduate studies in the discipline)? I'll admit I have no clear sense one way or the other and have no dog in the fight. But I do wonder whether there aren't some science writers who have attempted to survey what we know about human memory, or even some psychologists who set their research aside long enough to write up a general work. If you wanted to go further down this line of analysis, that might be an easy way to build on the initial hit in the Atlantic. Tangentially, in my own view, U.S. historians tend to be skittish--or is it evasive?--about whether there is some kind--any kind--of human nature that matters to how history has unfolded.  If that's of any interest, I can recommend Robert Sapolsky's Behave, although admittedly it would take you far afield and I have to note my own lack of expertise on the book's general subject. I think a key question you might want to ask is how exactly the problematic nature of Livermore's writing matters to the argument you are making in your book. 

Best wishes as you continue to work on the project, I've enjoyed learning about it! 

Best wishes, 

Dave Prior


Hi Lois,

Thank you so much for your comments. I appreciate not only the recommendation but also the perspective that you developed in dealing with a somewhat different figure. Your response brought several thoughts to mind.

First, in looking at Livermore’s account, I’ve been somewhat luckier in that I’ve been able to check it against a variety of other sources. Although he sometimes gets small facts wrong (e.g. the distance of a march), he is generally on the mark. Whether or not his impressions are valid (e.g. the nature of his relationships with his superior officers) is an altogether different matter.

Second, for the sake of brevity, I neglected to mention that Livermore did not wish to release his manuscript during his life for fear of antagonizing or embarrassing those he criticized. And even though the work was published in 1920, two years after his death, a number of names were excised from the book. The original manuscript as well as a transcription (both of which are held in the University of New Hampshire’s Library in the Special Collections) does name names. The foregoing may suggest to some that Livermore wrote with one eye looking over his shoulder and muted his observations to avoid trouble. I’d argue, however, that he took the precautions he did precisely because so many of his comments in the book were piquant and unrestrained.

Third (and I think the following goes to David’s point), one matter that does give me pause in referring to The Atlantic article is the idea that Americans in the middle of the 19th century may not have formed memories in quite the same way we do today. I know this issue has come up in discussions regarding Civil War soldiers and psychological trauma: can we really say that these men suffered from PTSD much like fighting men and women do in the 21st century?

Livermore’s account is problematic in the same way that many sources are problematic—but his work has a few unique wrinkles of its own. I like (and need) to think about the degree to which I can rely on him.



Dear David,

I think your opening questions get to the nub of a very important issue that I touched upon in my response to Lois:

“Does the Atlantic article suggest to us that the field of psychology has reached a broad consensus about what human memory is and how it operates, and that historians can draw on this consensus profitably (without each of us having to go through graduate studies in the discipline)?”

The answer to the first part seems to be, “yes,” and to the second part, “maybe.” The Atlantic appears to claim there is a consensus about memory among leading psychologists who study that field. Having said that, though, I do wonder if “historians can draw on this consensus profitably.” My main reservation is that psychologists have reached their conclusions based on studies of contemporary people. Do we know for a fact that folks 160 years ago constructed memories in the same way? There are certain elements of Livermore’s experience that suggest they did, but frankly, it’s hard to know for sure.

Thank you for the recommendations. In my experience, I’ve found that readings that appear far afield can be really useful in surprising ways. The only way to know more is to do and read more.



Hi Hugh and David,

One complicating factor as we ponder whether the way folks formed memories in the mid-19th century differed from the way they/we do so today is that we have far more media that record things we incorporate as memory: photographs, moving images, sound recording. Do I remember the sound of so-and-so's voice, or do I remember how that person sounded on the radio, on my answering machine, on a voice memo? It is abundantly clear to me that many of my 'childhood memories' are really memories of photographs or home movies I look at years ago yet also years after they were created; what would I 'remember' if I had not had those interim prompts?

Well, at least Hugh isn't having to rely on my memoirs of the Civil War . . .


Hi Lois,

Funny you should mention this multitude of media. I’m currently reading through Colonel Edward Cross’s journal (he was the first commander of the 5th New Hampshire and was mortally wounded at Gettysburg while commanding the brigade to which that regiment belonged). I’m only just discovering that after Cross’s death, Thomas Livermore also read the journal and inserted notes here and there, clarifying or correcting the colonel’s impressions. While this is not exactly the same thing as what you’re referring to, it does make me realize how many “prompts” Livermore subjected himself to as he wrote his account of the war.

At some point—whether it was before or after Livermore is unclear—William Child, who finished the war as the 5th New Hampshire’s surgeon and later wrote the regiment’s “official” history, also read the journal, presumably before he completed his book. It really does seem to be the case that all these veterans were engaged in constructing a collective memory of the war. . . .



Dear All, 

One more quick comment... I definitely concur with Lois's point that the prevalence of media, esp. audiovisual recordings, raise questions about how we remember, or think we remember, things vs. how people did in the past.  I will, however, add a quick caveat. One of the things that provides coherence to American popular culture and politics across the 19th, 20th, and 21st centuries is media saturation (for the CWE, think of works like David Henkin's City Reading and Mark W. Summers's The Press Gang, among many others). The kinds of media have changed, but not their superabundance and tendency to proliferation. Livermore's age was also the first great age of the illustrated press, which covered  and recreated, with various degrees of accuracy, Civil War battles ad nauseum. Just in case Hugh was looking for one more layer of complexity to add to his already richly challenging project. 

I do find myself wondering whether it is the case that historians, or at least those of the CWE, already have a fairly sophisticated sense of why and how humans misperceive and misremember events, or whether we just default to a generic sense of doubt about primary source reliability.  I'll humbly admit my own lack of clarity on the issue suggest I do the latter, no doubt others here are more expert. Hugh, this is completely tangential, but I do remember some interesting discussion in Michael Lewis's popular The Undoing Project about the specific ways in which people, when imagining a sequence of events playing out differently than they did, tend to do so in very specific ways. That is its own interesting angle. 

Best wishes, 


Hi Dave,

This is a fantastic point. I know from my research on my previous book that the 1860s witnessed the rise of the penny press in Britain and the explosion of newspaper circulations into the hundreds of thousands. My impression is that the leading American papers had even larger circulations during this period.

Thanks for the citations. The more I learn, the more I learn that I have so much left to learn.